Histone deacetylases regulate chromatin structure, transcription, replication, and DNA repair/recombination. Because of these key roles in cell cycle control, HDACs have become an important target for therapeutic intervention in a variety of tumors and the first Histone Deacetylase Inhibitor (HDI) has received FDA approval. Histone deacetylase 3 (Hdac3) is a key regulator of chromatin structure and gene expression and our preliminary data suggests that it may be a central target of HDIs. Because of the multi-faceted action of histone deacetylases, it has been difficult to pin down how inhibiting these enzymes acts to kill cancer cells while sparing surrounding normal tissues. Therefore, we set out to understand the action of Hdac3 by engineering mice containing a floxed Hdac3. Deletion of Hdac3 in the germ line caused early embryonic lethality. Moreover, Hdac3 deletion was even lethal to murine embryonic fibroblasts in vitro. Although we approached this analysis with a transcription bias, gene expression array analysis failed to identify the induction of a gene expression pattern that might explain this apoptosis. By contrast, careful analysis of the cell cycle indicated that apoptosis required S phase progression and that S phase was slowed. Surprisingly, the S phase DNA damage checkpoint was activated in the absence of Hdac3. This observation, coupled with data linking Hdac3 to DNA repair, led us to the discovery that loss of Hdac3 caused DNA damage that was most likely due to inefficient repair of the DNA double strand breaks that occur at stalled DNA replication forks. We have now begun to translate this information to the in vivo requirements for Hdac3 in adult mice. Removal of Hdac3 in the mouse liver caused a phenotype similar to non-alcoholic fatty liver disease, which was likely due to de- regulated nuclear hormone receptor functions. Most importantly, consistent with defects in DNA repair and genomic instability, the liver-specific deletion of Hdac3 caused a 100% penetrant hepatocellular carcinoma with a mean time to HCC of 10.2 months. We hypothesize that Hdac3 is required for efficient DNA repair and that loss of Hdac3 causes genomic instability that leads to cancer development, but that we can take advantage of these defects in DNA repair to treat cancer. This proposal will further define the mechanisms by which inactivation of Hdac3 affects DNA repair, regulates chromatin structure, and causes hepatocellular carcinoma.